---
language:
- zh
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_13_0
metrics:
- wer
model-index:
- name: Whisper medium zh - seiching
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: Common Voice 13
      type: mozilla-foundation/common_voice_13_0
      config: zh-TW
      split: test
      args: zh-TW
    metrics:
    - name: Wer
      type: wer
      value: 37.69215412257936
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# Whisper medium zh - seiching

This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Common Voice 13 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2083
- Wer Ortho: 37.9482
- Wer: 37.6922

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 50
- training_steps: 4000

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer     |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:-------:|
| 0.1472        | 0.69  | 500  | 0.1579          | 36.6425   | 36.4544 |
| 0.0545        | 1.38  | 1000 | 0.1685          | 37.4093   | 37.4725 |
| 0.0227        | 2.06  | 1500 | 0.1751          | 37.5544   | 37.9118 |
| 0.0262        | 2.75  | 2000 | 0.1885          | 37.9689   | 37.4925 |
| 0.0203        | 3.44  | 2500 | 0.2042          | 37.2228   | 36.7938 |
| 0.0123        | 4.13  | 3000 | 0.2065          | 38.3834   | 37.9916 |
| 0.0121        | 4.81  | 3500 | 0.2065          | 37.6373   | 37.7720 |
| 0.0151        | 5.5   | 4000 | 0.2083          | 37.9482   | 37.6922 |


### Framework versions

- Transformers 4.30.2
- Pytorch 1.13.1+cu117
- Datasets 2.13.2
- Tokenizers 0.13.3